Exchange on Bruce Waller's
Against Moral Responsibility
Clark comments on Dennett's review of Against Moral Responsibility
Thanks for your review of Waller; it’s forceful, well-written and makes some very fair points. You’re generous in your praise, for instance in crediting him with getting you to better articulate your own position. It’s also very helpful in setting forth your consequentialism and non-retributivism, which I think supports the project of creating a less punitive culture, even if your main focus is on defending punishment. And despite your disagreements with Waller, there’s considerable progressive common ground. Here are some further comments, pretty much in order as they come up in the review:
You say Waller makes an original mistake, that he asserts a heretofore unexamined falsehood. After reading the review I still wasn’t sure what this was, unless you’re referring to his overall thesis against moral responsibility (MR).
You say that “Almost everybody wants to secure a justification for retributive blame and punishment, just deserts…” But it seems you are not among them, since you espouse a reformed, non-retributivist, consequentialist concept of MR and punishment. I’m hoping that a sizeable minority would agree with you and Waller (and Greene and Cohen, among others) that we should dispatch retributive punishment and move to a purely consequentialist system. But for some reason you want to keep “just deserts” in the lexicon, which I think is inconsistent with your reformed concept of MR.
“Waller opts at the outset for a strong definition of moral responsibility…” Waller’s definition of MR, tying it to the retributivist, non-consequentialist conception of desert, is canonical. I point this out in at the start of my review of Waller, and could cite more examples, e.g., compatibilists Stephen Morse and Michael Moore both are explicitly deontological and retributivist about desert. If compatibilists have indeed abandoned this “unreformed, unsophisticated” notion of MR and are moving en masse to secure a “suitable substitute,” that’s news to me and it’s good news indeed that they are abandoning retributivism. But unfortunately I think you are very much in the compatibilist vanguard in embracing consequentialism; let’s hope others like Morse and Moore come to see the light.
“A consequentialist defense of just deserts….” Whether as consequentialists we should still talk of just deserts is debatable, given the strong deontological, retributive connotations (parallel to the debate about whether we should still say we have free will given its unsustainable libertarian connotations). What you’re advocating is the practical necessity of punishment, not its intrinsic goodness, but “just deserts” strongly implies that the offender’s suffering is intrinsically good, which you don’t think is the case. So I think we should drop talk of just deserts so we don’t mislead people about what we believe are defensible justifications for punishment.
“…you have to start with the recognition that the concept of moral responsibility [MR] doesn’t just drop into our conceptual scheme from the sky; it is integral to a large system of social and political institutions…” I agree, and would add that the concept is also strongly influenced, for better or worse, by our innate reactive dispositions (hence not entirely artefactual) and a long history of belief in contra-causal free will (hence non-naturalistic). Further, MR ultimately gets cashed out in terms of our responsibility practices and how we justify them. But none of this means that the concept of MR can’t be questioned in the fundamental way Waller recommends. Our current responsibility practices (e.g., death penalty, solitary confinement, tolerated rape in prison), strongly influenced by folk intuitions that we are indeed “guilty-in-the-eyes-of God,” help to define MR: what people deserve, and why. All of this is up for grabs, as exampled by your condemnation of “our American system of justice and punishment [which] is obscenely mis-instituted for the most part.” So the anchoring of MR “in the (naturalistic) land of ethical theory, economics, political theory, psychology and other social sciences” doesn’t protect it from radical revision or perhaps even elimination (what Waller and other MR skeptics recommend), if we judge that dropping deontological desert and retributive punishment changes the concept beyond useful recognition.
“I agree with Waller’s main conclusion in one important sense: that kind of absolutistic moral responsibility—insisting as it does on what I have called guilt-in-the-eyes-of God—is incompatible with naturalism and has got to go. Good riddance.” This is encouraging common ground, and what Waller and other naturalist revolutionaries are trying to do is show the humanizing consequences of debunking libertarian free will and ultimate responsibility, for instance in criminal justice reform and reducing inequality. You spend little time on this, but instead focus on defending punishment against those who you see as wanting to abandon it.
“He, too (I imagine—he never mentions it), wants to maintain promising and contracts and the institutional understandings that make them possible, but if so, I guess he doesn’t view the penalties and justification of taking-by-force-if-necessary of these institutions as punishment (real, retributive punishment), since after all, they have a consequentialist grounding, as just articulated. To make the contrast clear, that concept is my “core concept” of moral responsibility, an ultimately consequentialist, not retributive concept.” This is more common ground between you and Waller, except that he wants to drop MR altogether, whereas your concern is to point out the necessity of punishment as a bulwark against bad behavior, which I don’t think anyone is contesting. And after all, people love to punish, they find it rewarding (as B.F. Skinner pointed out), so it isn’t as if punishment needs apologists. It’s limiting our appetite for and reducing the need for punishment, seems to me, that should primarily engage us. And we can help that cause by publicly debunking libertarian free will and ultimate responsibility as justifications for and incitements to retribution.
“Waller gives us example after example in which he contrasts two people, one of whom had a salubrious upbringing while the other was deprived of most benefits. Is it fair, he keeps asking, to hold both of them responsible?” What Waller says isn’t fair is to suppose that the fortunate deserve all the rewards that come from being lucky in life, or that the unfortunate deserve all the punishment or lack of rewards that come from being unlucky. We have to hold people (consequentially) responsible to help keep them in line, but dropping the concept of just deserts will help make our responsibility practices and social policies more humane and effective, producing more equitable opportunities and outcomes.
“Think about sports…” You’re using the game analogy to defend the fairness of assigning just about everyone the status of a moral agent, whereas Waller uses it to point out the unfairness of traditional desert. Games are primarily about ensuring fair competition between more or less equally talented players to see who’s best; we don’t put novices up against experts and expect a good game. In games everyone starts off even at the start, and at the beginning of each season everything’s reset to zero. In life (“gnarly reality”) nothing is ever reset, people of vastly different competencies are in competition, and despite the best efforts of liberals, the playing field remains drastically tilted toward those with original and accumulated advantages. As you say (and I think Waller would agree), as a matter of social necessity people of vastly different competencies and backgrounds all count as a sanctionable moral agents: as a matter of policy just about everyone meets a somewhat arbitrary and adjustable criterion for being held responsible. Fine. But Waller’s point is that MR does nothing to even the playing field of life, rather quite the opposite. It says that the winners and losers in life essentially deserve what they get, so inequality is fair - don’t look too hard at the system or at what caused a loser to end up that way. You seem to agree, even on your reformed version of MR, since you say luck evens out in the long run, giving the initially disadvantaged an equal chance at success. So if someone doesn’t succeed it’s still mostly their fault; it’s fair and efficacious to primarily blame them, not the system which produced them. But Waller and I and others disagree. We can do better by not singling out the agent; instead, publicize naturalism-determinism, which undercuts ultimate responsibility as a justification for both punishment and inequality. Your focus, however, is on defending punishment and the basic fairness of the current system, not on promoting the naturalistic revolution in our self-concept.
“But I want to reform, not abandon it…” I don’t think Waller or anyone else is suggesting we abandon our criminal justice system or the “omnipresent threat of punishment,” for instance to keep greedy doctors in line. As you point out, “Waller concedes that it [blame] does work to a considerable degree.” And he would agree that compatibilist criteria rightly distinguish the guilty from innocent; there’s no get out of jail free card on his view – he abjures what he calls “excuse extensionism.” It’s just that he’s pointing out the unfairness and excesses of retributive punishment based in unreformed MR (solitary confinement and eventual insanity for Madoff) and you’re emphasizing the role of punishment as deterrence.
Your hard-nosed challenge to Waller and other reformers to get real is well-taken. What would a criminal justice system look like in light of consequentialism? We need to come up with a program of reform motivated by a naturalistic critique of the “indefensible concept of moral responsibility” that many of the folk likely hold. The barrage of rhetorical questions you pose about Madoff (a bit of piling on?) need serious discussion in fleshing out what a less punitive, more effective system, necessarily embedded in a more enlightened culture, would look like; some Scandinavian countries come to mind. We all want to avoid the ominous threats to human rights you (wrongly, I think) accuse Skinner of abetting, and of course we want to reduce the incidence of wrong-doing in the first place, what Waller’s emphasis on the causal and situational determinants of crime encourages. To say, as you do, “A world without punishment is not a world any of us would want to live in” obscures all this and makes it sound as if punishment is an intrinsic good (as retributivists would have it), not an unfortunate practical necessity. On the other hand, to maintain that it isn’t currently a necessity is pie in the sky, so if Waller is guilty of that he stands corrected.
“It is open to Waller… to propose a better game…” Waller highlights the causal story of those who are unlucky in life in order to promote a preventive, ameliorative approach to deviance, dysfunction and inequality, whereas you and other compatibilist apologists for punishment focus on maintaining threat-based deterrents to bad behavior or defending retribution itself (as do Morse and Moore). Compatibilists abet the propensity to punish and ignore causes, it seems to me, by downplaying determinism and the systems view while focusing attention on the “sins of the error-maker.” When it comes to creating a less punitive, more humane culture, it’s Waller’s approach we most need, the one that puts agents explicitly in their causal context: no one has the unconditional ability to have done otherwise. We can best improve on the state of nature by advertising, not hiding, the naturalistic revolution in our self-concept.
- TWC, October, 2012
I thank Tom for his constructive response to my review, and for vigorously drawing my attention to Bruce Waller’s book in the first place. And of course I also thank him for hosting this naturalism website, a valuable resource to which he himself has contributed so much. There is a lot we agree about, as he notes, but there remain some points of disagreement that I hope we can soon resolve.
First, Tom wonders just what mistake I think Waller is making. It is, as he supposes, given by Waller’s title; he is against moral responsibility, and this, I think, is a big mistake, made possible by a smaller mistake: equating moral responsibility (and its companion, just deserts) with its extreme, retributive version. It is as if he thought punitive was synonymous with retributive. It is not. There is a perfectly good consequentialist justification for society’s punitive measures (when properly administered), just as there are perfectly good consequential justifications for penalties in sports. They make the games possible, and that is a good consequence.
Tom disagrees, claiming that Waller’s definition of just deserts is ‘canonical’, not extreme, and that the received view of punishment is retributive through and through. I’m not sure how to settle this disagreement. Certainly counting noses, (corrected, if you like, for differences in gravitas or eloquence) is not the way. So I’ll concede the point, not because it then puts me, as Tom says, in “the vanguard,” but for the sake of argument. Let’s suppose he is right, for the moment. Tom goes on:
“Strongly implies”? Isn’t that an oxymoron? Either just deserts implies that the offender’s suffering is intrinsically good or it doesn’t. I think what Tom means is that it strongly connotes that the offender’s suffering is intrinsically good, and this might be true, but I doubt it, and I also question its importance here. Is there a big difference between something that is “intrinsically” good and something that is a practically necessary condition for something that is intrinsically good? The only way I can see of securing civilization (which I see as about as “intrinsically” good as anything I can think of) is with a system of moral responsibility that includes punishment for those found responsible for their misdeeds. Is food and water intrinsically good? Or are they “just” requirements for life (which is “intrinsically good”)? Food and water are practically necessary; so is punishment. The utopian idea that we might reform human nature so much that punishment was no longer practically necessary is about as realistic as the idea that we could reform human biology so much that food and water would no longer be a “practical” necessity for human life. (Why might one want to change human nature so that food and water weren’t necessary? Well, then we could go hiking for months on end without carrying our own supplies, or sail around the world in a small boat with no concern for provisioning. Wouldn’t that be nice?) And never forget that the other side of the moral responsibility coin is reward; do we really want to live in a world where great contributions to humanity are not rewarded? Those who argue passionately for ridding the world of moral blame are notably tongue-tied when it comes to impressing us with the urgency of dismantling our practices of honoring and rewarding our heroes, but consistency demands that praise and blame stand or fall together. Given the way human nature is now, and will be for the foreseeable future (barring technological “miracles” that turn us all into saints), praise and blame, reward and punishment, are as necessary for civilization as food and water are for life. That doesn’t make punishing the guilty “intrinsically good” in the extreme sense that Kant defended, but it does make it a very important good, a practically necessary good, a staff of life.
Now let me set aside my concession and argue against the claim that retributivism is canonical—not just the default presumption of folks who haven’t given it much thought, but firmly in place in the foundations of reflective people’s thinking about punishment. Consider Kant’s notorious view that we should execute those on death row on the eve of the end of the world. This is a definitive statement of canonical retributivism if anything is. If you think punishment of the guilty is intrinsically good, how could you deny it? Kant’s formula nicely strips away all possible consequentialist options, leaving the intrinsic good of such punishment as the only end in sight. But I suspect almost nobody endorses it today, so perhaps there are fewer pure retributivists around than Clark and Waller suppose. Can they point to any retributivists who bite the Kantian bullet?
We can all feel the grip of the retributivist urge. Consider watching a movie that ends with the horrible villain escaping his pursuers and sailing off to his private island well stocked with all life’s pleasures. Not a happy ending! Now rewrite the ending; imagine that his triumphant smile suddenly turns into a rictus of horror as a great white shark rises up to snatch him from his yacht and make a meal of him. YESSS! Clark and Waller are probably right that this sentiment is built right into our genetically evolved sociality, but they need to show that it hasn’t been—and can’t be—harnessed and tamed by our socially evolved institutions of praise and blame, reward and punishment, turned, as Hume would put it, from a natural vice into an artificial virtue. One might well respond to Kant’s gruesome scenario by admitting that while it would probably be intensely satisfying to watch the death row villains endure some extra suffering before the curtain came down on existence, there would be no duty to bring it about, and even a duty to prevent others from bringing it about.
I agree completely with Tom that anchoring moral responsibility in the naturalist world doesn’t protect it from criticism of the sort Waller attempts. That is what makes his book so valuable: he finds novel ways of challenging what has largely gone unchallenged. These novel ways just don’t succeed. Tom says I spend little time developing a suite of reforms of the criminal justice system, and that is true, but mainly because I think the reforms most pressingly needed are obvious: a drastic diminution in the length of sentences, and indeed the elimination of most incarceration in favor of less drastic penalties, much better enforcement of prisoners’ rights, and better programs for reintegrating prisoners into society. (It is sobering to reflect that the penal system most white collar criminals face—think of Martha Stewart—is a pretty good model for every kind of criminal incarceration.)
Tom sees his mission as using naturalism as an authoritative banner to rally those of us who abhor our current punitive practices, but by my way of thinking, his (and Waller’s and Greene and Cohen’s) campaign does a disservice by distracting attention from reform by calling for abolition, an obviously unrealistic and even undesirable utopian alternative that people know in their bones is hopeless. Nothing in naturalism favors abolition over reform. It is that simple. Now let’s stand shoulder to shoulder for reform.
Tom says “dropping the concept of just deserts will help make our responsibility practices and social policies more humane and effective.” I disagree. As I point out in my review, the maintenance of just deserts provides a bracing environment in which parents are strongly encouraged to devote time and effort to moral education—to cite just one such effect—in much the same way as strict liability laws provide a bracing incentive to those they cover. There is now mounting empirical evidence (beginning with Vohs and Schooler, but extending well beyond their pioneering experiment) showing that a belief that free will is an illusion (something reported as following from naturalism) encourages antisocial behavior. So there is actually good empirical evidence against Tom’s claim that our social policies will be “more effective” if we pursue his goals. Tom also says:
This is two points, and neither has been shown, not remotely. First, we have to “single out the agent” in some regards just to identify the problem, and, in some cases, to protect society from the agent while we figure out what to do. Nobody, I take it, is urging us to stop investigating crimes to find the perpetrators. So like it or not, perpetrators will be identified. And like it or not, their mere identification will already add to their life’s burden, which may indeed be a product of disadvantages and hardships. Tom needs to come to grips with this point: suppose he is mugged, and is asked to identify the mugger in a line-up, and suppose he can do so with great confidence. Should he? It would seem he should be a conscientious objector and refuse to cooperate with the authorities, since their subsequent actions will be unfair, no matter what they are, since, according to Tom, and to Waller, the mugger is not morally responsible for the mugging and hence should not be singled out. That, Tom says, is the moral to draw from naturalism-determinism. I doubt if Tom, or Waller, wants to live in that world. I know that I do not.
So suppose Tom grants that we may—and ought to—single out the mugger when we can do so with reliability, but we must do everything in our power to prevent the mugger from being blamed. This gets delicate: we can forcibly quarantine the mugger, forcibly enroll the mugger in rehabilitation programs, go out of our way to warn the citizenry that this individual is probably dangerous and should be given a wide berth—but we must not blame the mugger. One begins to wonder if we’re just banishing a word while keeping all the traditional implications of the word intact. And how long may we quarantine the mugger? As short a period of time as is consistent with public safety. The mugger will, of course, plead that he has already learned his lesson and is ready for release, but how will we assess this? A curious feature of the general policy Waller and Clark—and others—propose is that the very uncertainties that can cloud the backward-looking judgment of desert also cloud the forward-looking judgment of rehabilitation, a point forcefully made by Stephen White (Tufts symposium on Does anybody ever deserve anything?, October 12, 2012). Note that the burdens of proof are reversed: the protestation of incapacity (“I could not have done otherwise!”) that absolves from blame must turn into a protestation of newfound capacity (“I can in the future do otherwise!”). But if “neuroscience proves” that people’s intuitive judgments of capacity are all illusory, we have no reason to give such protestations any weight. Face with a choice between enduring a fixed punishment followed by release and restoration of rights and the prospect of indefinite quarantine and rehabilitation, few would discard their moral responsibility gratefully.
And note, finally, that naturalism-determinism doesn’t undermine what Waller calls take-charge responsibility; it only undermines ultimate responsibility, but that puffed up notion is not necessary as a foundation for a justifiable, fair, reasonable institution of reward and punishment, praise and blame. There is no reasonable alternative to naturalism, but naturalism does not have the implications Tom Clark and Bruce Waller think it does. If they still think otherwise, it falls to them to come up with an argument that shows this.
- DCD, October, 2012
My sincere thanks to Tom Clark for setting up this discussion, for his outstanding website, and for his vigorous defense of our common views; to Daniel Dennett, for his generous, incisive, and absolutely fair response to Against Moral Responsibility; and to both Tom and Dan for the many insights I have gained from their work. There is no book I have enjoyed more or learned more from than Dennett’s Elbow Room. Dennett’s review has given me a much clearer picture of his position, as well as a keener appreciation of the vulnerabilities of my own views. One of the aspects of Dennett’s work that is particularly impressive is the seriousness with which he takes philosophical questions. That deep seriousness certainly does not make his work somber – no philosopher writes with a lighter touch or is more deeply amusing than Dennett (who else would start a very important philosophical work with an admonition not to feed the bugbears?) – but the fact that Dennett takes philosophical issues very seriously indeed shines through even his most amusing passages. That is a characteristic that Dennett’s work shares with Clark’s; and I share with them the conviction that questions concerning moral responsibility are fascinating philosophical questions, but also questions with profoundly important social implications. Also, it seems to me that Dennett’s review is a model of clarity and fairness: he strongly disagrees with my conclusions, yet he is scrupulously fair in presenting my views and my arguments. (Tom often presents my views better than I can, but that is not so remarkable: he has thought about these issues long and carefully and productively, and we have very similar views.)
There is much in Dennett’s review with which I fervently agree. First, I would be proud and delighted to “stand shoulder to shoulder for reform” with Dan and Tom. We all agree, and Dennett states that agreement quite forcefully, that “our American system of justice and punishment is obscenely mis-instituted.” As Dennett notes, he wants to reform it while I ultimately want to abandon it; but in the interim, it would be a wonderful improvement if we could move toward a genuinely consequentialist system that would make a sincere effort to devise institutions and programs that would “even the playing field as best we can, by minimizing the amplification of advantage and disadvantage that otherwise would probably occur.” Indeed, Dennett suggests that we might look upon the treatment of Martha Stewart as a criminal justice ideal. If I live to see the day when the U.S. makes a genuine commitment to a more egalitarian society, abolishes capital punishment, destroys all its Supermax prisons and reduces the number of prisoners to a minimal number with those few being treated much as Martha Stewart was treated during her period of “imprisonment” (all of which Dan seems to endorse) then I would be a very happy camper indeed. But our points of agreement notwithstanding, now we reach some points of disagreement. First, all three of us want to travel in the direction of major reform; though Tom ultimately wants to go further than Dan, and I want to pursue that path all the way to a point at which Dan would certainly insist (and even Tom fears) I plunge directly over a steep cliff. But Tom and I both believe that commitment to moral responsibility is an impediment to such a reform agenda, and at that point we get to some basic questions.
The fundamental question at issue is whether our practice of holding people morally responsible is fair. That is a question Dennett very carefully examines, and he rightly insists that it is a question distinct from the obvious unfairness of our grossly excessive and counterproductive punitive system. I won’t repeat his arguments here: they are wonderfully clear and powerful as he presents them in his review. I hope I am not misreading his conclusion, but it seems to me that Dennett does not argue that our moral responsibility system is fair; rather, it is fair enough, and – though we ought to continue to strive to make the system as fair as it can possibly be – the system of moral responsibility (even with some modest degree of unfairness) is far better than any system we could put in its place (and certainly better than any system I have suggested as a replacement). Dennett seems comfortable with “fair enough,” and he can champion such a system and not blink. But Dennett is a remarkable person, and what works for Dennett is unlikely to work for most people. The idea that we are benefitting and others are suffering from a system that is not quite fair creates “cognitive dissonance” for most people. We resolve it by convincing ourselves that the system really is fair, and people justly deserve what they get; and the best way of convincing ourselves of that is by refusing to look in depth at the detailed causes for differences in behavior (for example, the causes that shape one person as a cognitive miser and another as a chronic cognizer, than enhance one person’s sense of self-efficacy while diminishing another’s). As a result, the real problems in the system tend to be invisible, and the deeper understanding of human behavior -- that contemporary psychological research is now offering, and that affords us a genuine opportunity to enhance the self-control and freedom of everyone – is neglected.
In fact, I’m not sure that even Dennett is quite comfortable with the unfairness – even under optimum conditions – of the moral responsibility system. He adds another prop to his argument: The system is fair enough because we choose to participate and play the game, and so we must regard it as at least reasonably fair. But here the sports/game metaphor becomes less helpful. If I don’t like football and its rules, I can play a new game. If athletic games don’t answer, I can turn to backgammon. And if I really want something where my athletic ineptitude and gaming incompetence are not a problem, there’s always chutes and ladders. But in life we can’t select another game, or choose a different social system: this is the only one available, unless we decide to be hermits. For profoundly social animals, that’s not an attractive option – especially when those who remain in the game would classify us as demented, and fit only for treatment, and unworthy of respect or affection. If we don’t play we are banished not just from society, but from the human community. (Given that alternative, it is hardly surprising when even those who are most severely mistreated -- by a U.S. system Dennett agrees is grossly unfair -- insist that they want to remain in the game.)
But Dennett saves his toughest challenge for last. So you don’t think this “fair enough” system is acceptable? Fine; show us a better one. Dennett works through the tangle of arguments and claims, and tracks down the most vulnerable point in the case against the moral responsibility system. It is because Dennett is wonderfully honest about the limitations of the moral responsibility system that his challenge has such force: OK, nobody claims that the moral responsibility system is perfect; but do you have anything better to put in its place? At this point (Dennett and I share a fondness for sports metaphors) I’m backed up to my own goal, it’s 4th and 10, and I’m tempted to punt. Or maybe, inspired by Dennett’s example, I should be honest: I don’t have such a system to offer. But on the other hand, I don’t think that is quite so severe a problem as Dennett implies; in fact, I don’t think it is really that surprising. Formulating such a system is not primarily a task for philosophers, though we can help. But in any case, Dennett is right: no such alternative is now available. He is also right that some of the new developments (that I regard as promising, such as restorative justice) now rely on a background punitive possibility. But we are now taking small steps, experimental steps, in the right direction; I regard those steps as somewhat more promising than Dan does, but I agree with him that they are not a full alternative to the moral responsibility system.
I don’t have an alternative; but my goal is to convince the world (starting with philosophers, the most important part of the world) that the moral responsibility system has inherent flaws, that it is unfair, and that it blocks a better understanding of human behavior. If we acknowledge that the system is unfair, then we – psychologists, criminologists, sociologists, philosophers, and candlestick makers -- can and will work toward something new which makes good use of all we now know about human behavior (and that does not mean, nor would B. F. Skinner have supposed it to mean, turning people into automatons: a very important part of such a project will involve using our new understanding to enhance self-control, strengthen a sense of self-efficacy, and promote stronger cognitive resources). Is that possible? I don’t know; my hope that it is possible is not “pie in the sky,” but is based on the fact that we have made enormous gains in our understanding of human behavior, and are constantly making more, and we have barely begun to make effective use of that understanding. But the first step is to be clear that it’s worth trying.
~ BNW, October, 2012
[back to Reviews]